EGU24-5103, updated on 08 Mar 2024
EGU General Assembly 2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

Reconstructing Historical Climate Fields With Deep Learning

Nils Bochow1,2,3, Anna Poltronieri1, Martin Rypdal1, and Niklas Boers3,4,5
Nils Bochow et al.
  • 1UiT The Arctic University of Norway, Faculty of Science and Technology, Department of Mathematics and Statistics, Tromsø, Norway (
  • 2Physics of Ice, Climate and Earth, Niels Bohr Institute, University of Copenhagen, Copenhagen, Denmark
  • 3Potsdam Institiute for Climate Impact Research, Potsdam, Germany
  • 4Earth System Modelling, School of Engineering & Design, Technical University of Munich, Munich, Germany
  • 5Department of Mathematics and Global Systems Institute, University of Exeter, Exeter, UK

Historical records of climate fields are often sparse due to missing measurements, especially before the introduction of large-scale satellite missions. Several statistical and model-based methods have been introduced to fill gaps and reconstruct historical records. Here, we employ a recently introduced deep-learning approach based on Fourier convolutions, trained on numerical climate model output, to reconstruct historical climate fields. Using this approach we are able to realistically reconstruct large and irregular areas of missing data, as well as reconstruct known historical events such as strong El Niño and La Niña with very little given information. Our method outperforms the widely used statistical kriging method as well as other recent machine learning approaches. The model generalizes to higher resolutions than the ones it was trained on and can be used on a variety of climate fields. Moreover, it allows inpainting of masks never seen before during the model training.

How to cite: Bochow, N., Poltronieri, A., Rypdal, M., and Boers, N.: Reconstructing Historical Climate Fields With Deep Learning, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5103,, 2024.